INDEX

Explanations

phrases and numerical measures related to performance metrics in a technical context

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AsUp

-0.54

 dévelo

-0.50

enderror

-0.50

 ainfi

-0.50

 OFDb

-0.50

 nonUne

-0.49

 Vikipedi

-0.49

 noDo

-0.49

CascadeType

-0.49

 individu

-0.48

POSITIVE LOGITS

 eikä

0.37

 muerta

0.35

BASELINE

0.35

 enamorado

0.34

roaches

0.34

IntoConstraints

0.34

MessageTagHelper

0.33

hmig

0.33

 ontem

0.33

DATABASES

0.32

Activations Density 0.045%

phrases and numerical measures related to performance metrics in a technical context

No Comments

No Known Activations